Many of these studies are fraught with problems. The one that researched SF congestion for example almost perfectly coincides with the multi-year construction o

Many of these studies are fraught with problems. The one that researched SF congestion for example almost perfectly coincides with the multi-year construction on Van Ness completely disrupting north-south traffic in the city. You also have more and more people in every growing city supercommuting due to rising housing costs. A lot of these supercommuters are driving their own personal vehicle because they are now living in places where a personal vehicle is a necessity. Once you have your own personal vehicle for local use, it's economically advantageous to use it for commuting to work. Given all the changes in growing cities in the past 5-10 years, you couldn't even begin to isolate the cause of congestion down to just one cause, but that's what these studies falsely claim because they don't understand the null hypothesis and that correlation does not imply causation.